Benchmarks · knowledge
Harvey LAB-AA v1.1
Harvey LAB-AA v1.1: 120 private agentic legal tasks built by Harvey. Hallucination-Gated All-Pass Rate: a task counts only when a three-judge panel passes every rubric criterion and the deliverables contain no material hallucination. Not comparable with v1.0 scores.
22 models · 26 results · data as of 2026-11-06 · artificialanalysis.ai ↗
Leaderboard
Muse Wonder (Max)
More from this benchmark
88 published measuresScore vs any published measure88 measures: sub-scores, splits, counts
11 more models under All models in settings.
Breakdown
darker is better within each column · click a column to sort| Model | |||||
|---|---|---|---|---|---|
| Muse Wonder (Max) | 0.228 | 0.3 | 0.892 | 1.26 | 117.0 |
| Grok 4.7 xhigh | 0.153 | 0.231 | 0.742 | 0.8 | 63.4 |
| Muse Spark 1.3 max | 0.144 | 0.189 | 1.68 | 1.91 | 93.8 |
| GPT-6 Astra max | 0.203 | 0.317 | 0.0333 | 0.075 | 54.3 |
| GPT-6.1 Sol max | 0.203 | 0.296 | 0.0917 | 0.175 | 56.9 |
| Fable 5.1 max | 0.122 | 0.164 | 1.32 | 1 | 83.2 |
| Kimi K3 max | 0.075 | 0.0972 | 2.09 | 1.8 | 84.4 |
| Opus 5.5 max | 0.0917 | 0.161 | 0.592 | 0.508 | 116.9 |
| Gemini 4 Argon high | 0.0694 | 0.0944 | 3.82 | 1.82 | 67.2 |
| GPT-6 Sol max | 0.1 | 0.181 | 0.0667 | 0.175 | 75.2 |
| GPT-6 Luna max | 0.0889 | 0.153 | 0.375 | 0.5 | 76.1 |
| Haiku 5.5 xhigh | 0.0528 | 0.0583 | 1.61 | 1.67 | 79.6 |
| Sonnet 5.5 max | 0.0917 | 0.169 | 1.03 | 0.942 | 179.3 |
| DeepSeek V4.1 Flash max | 0.0389 | 0.0639 | 2.58 | 2.02 | 105.5 |
| GLM 5.3 Flash max | 0.0139 | 0.0167 | 4.04 | 3.43 | 79.3 |
| GLM 5.3 max | 0.0194 | 0.0222 | 3.94 | 3.6 | 93.5 |
| Gemini 3.1 Pro | 0 | 0.00556 | 1.07 | 0.45 | 20.0 |
| Gemini 3.5 Flash-Lite high | 0 | 0 | 9.06 | 3.04 | 17.0 |
| Gemini 3.8 Flash high | 0 | 0 | 14.0 | 6.43 | 70.1 |
| Inkling Small | 0 | 0 | 7.24 | 4.83 | 31.4 |
| Inkling xhigh | 0 | 0 | 10.4 | 6.42 | 48.7 |
| MiniMax M3 | 0 | 0.00278 | 7.06 | 4.88 | 63.2 |
All results
26 rows| # | Model | Effort | Harvey LAB-AA v1.1 |
|---|---|---|---|
| 1 | Muse Wonder (Max) | max | 10.0% |
| 2 | Grok 4.7 | xhigh | 9.4% |
| 3 | Muse Spark 1.3 | max | 8.9% |
| 4 | GPT-6 Astra | max | 8.6% |
| 5 | GPT-6.1 Sol | max | 6.9% |
| 6 | Fable 5.1 | max | 6.4% |
| 7 | Kimi K3 | max | 5.3% |
| 8 | Opus 5.5 | max | 4.2% |
| 9 | Gemini 4 Argon | high | 3.9% |
| 10 | GPT-6 Sol | max | 3.6% |
| 11 | GPT-6 Luna | max | 3.3% |
| 12 | Haiku 5.5 | xhigh | 2.8% |
| 13 | Sonnet 5.5 | max | 2.8% |
| 14 | Haiku 5.5 | low | 1.7% |
| 15 | DeepSeek V4.1 Flash | max | 1.7% |
| 16 | Haiku 5.5 | high | 1.1% |
| 17 | GLM 5.3 Flash | max | 0.8% |
| 18 | Haiku 5.5 | max | 0.6% |
| 19 | Haiku 5.5 | medium | 0.6% |
| 20 | GLM 5.3 | max | 0.3% |
| 21 | Gemini 3.1 Pro | 0.0% | |
| 22 | Gemini 3.5 Flash-Lite | high | 0.0% |
| 23 | Gemini 3.8 Flash | high | 0.0% |
| 24 | Inkling Small | 0.0% | |
| 25 | Inkling | xhigh | 0.0% |
| 26 | MiniMax M3 | 0.0% |